FEOGARM: A Framework to Evaluate and Optimize Gesture Acquisition and Recognition Methods
IEEE International Conference on Systems, Man, and Cybernetics, Anchorage, Alaska, USA, (октября 2011)Аннотация
This paper discusses the general concept and implementation steps towards a framework facilitating the development and evaluation of gesture acquisition and recognition systems through a common benchmarking standard. In particular, a complete ground-truth of annotated gestures will be acquired with multiple high-resolution acquisition devices and as such will facilitate, using machine-learning techniques, the development of non-intrusive and light capture devices dedicated to gestures recognition. This article presents our concept, the current achievements and the first steps towards such a framework and corpus.